Progress in FY2015 was made in the following areas: (1) RAPID MIXING AND FREEZING TECHNOLOGY FOR SSNMR. We have designed microfluidic mixer and freeze-quench systems for millisecond-time-scale studies of protein folding, aggregation, and ligand binding processes by solid state NMR. As an initial test, we have examined the pH-dependent folding and tetramer assembly process of the bee venom peptide melittin, using DNP-enhanced solid state NMR measurements to determine the conformational state of the peptide in frozen solutions at low pH, at neutral pH, and within 5-10 ms of a pH jump from low to neutral. Initial data indicate that we will be able to probe intermediate states in this process. We expect to exploit the new apparatus in studies of several different protein systems in the coming year. (2) MRI MICROSCOPY. We have demonstrated that 3D images of test samples can be acquired with 5 micron isotropic resolution in the case of liquids and 8 micron isotropic resolution in the case of solids, using the microcoil-based MRI system described in our FY2014 report. These results are at room temperature. The liquid state performance is similar to the best results obtained to date by other groups. The solid state performance is substantially better than any previous results from other groups. This work is described in a paper that is under review at J. Magn. Reson. We are now testing the performance of our MRI apparatus at low temperatures (20 K or less), where we can further enhance NMR signals by dynamic nuclear polarization. We expect to achieve isotropic resolution of 1 micron or less at low temperatures, which will allow us to visualize internal structures within cells and cell cultures, a long-standing goal of magnetic resonance imaging. (3) SIGNAL ASSIGNMENTS IN MULTIDIMENSIONAL NMR. In previous years, we have developed new computational algorithms for assigning signals (i.e., chemical shifts) in multidimensional NMR spectra of proteins or other biopolymers, based on a very general Monte Carlo/simulated annealing approach. We have applied this approach in numerous projects within our research group. This year, we have performed a new computational study that further elucidates the dependence of the signal assignment process on NMR linewidths, protein secondary structure, and the availability of various types of 3D NMR data. The results of this study have been published in J. Magn. Reson.